Designing PSO-Based PI-type Fuzzy Logic Controllers: A Typical Application to Load-Frequency Control Strategy of an Interconnected Hydropower System

Abstract

Recently, a fuzzy logic technique - based control strategy has become one of the most efficient control schemes in dealing with complicated and nonlinear control systems. The working principle of a fuzzy logic controller (FLC) is based on experiences of experts, thereby it should be one of the artificial intelligence controllers which can efficiently replace with conventional regulators. This paper concentrates on the design of a typical PI-type FLC incorporation with the particle swarm optimization (PSO) technique. The PSO algorithm is used to successfully determine three scaling factors of the PI-type FLC, thereby the proposed control strategy will be able to obtain much better control performances in comparison with the conventional regulator and/or the FLCs applying the other optimization counterparts, i.e. genetic algorithm (GA) and differential evolution (DE) algorithm. This study also investigates feasible solutions for the load-frequency control (LFC) problem of an interconnected hydroelectric power system. In fact, the LFC problem is one of the most important control issues of multi-area interconnected hydroelectric power systems. Comparative simulations with conventional PID regulators, PD-type FLC, GA- and DE-based PI-like FLCs will be implemented in this study to demonstrate the effectiveness of the proposed PSO-based PI-type FLC. The control quality of the proposed control approach revealing through major control criteria obtained, such as overshoots, steady-state error and settling time, is much better than that of the other controllers for solving the LFC problem of a hydropower system.